Coverage for /pythoncovmergedfiles/medio/medio/usr/local/lib/python3.10/site-packages/dill/__info__.py: 80%
Shortcuts on this page
r m x toggle line displays
j k next/prev highlighted chunk
0 (zero) top of page
1 (one) first highlighted chunk
1#!/usr/bin/env python
2#
3# Author: Mike McKerns (mmckerns @caltech and @uqfoundation)
4# Copyright (c) 2025 The Uncertainty Quantification Foundation.
5# License: 3-clause BSD. The full license text is available at:
6# - https://github.com/uqfoundation/dill/blob/master/LICENSE
7'''
8-----------------------------
9dill: serialize all of Python
10-----------------------------
12About Dill
13==========
15``dill`` extends Python's ``pickle`` module for serializing and de-serializing
16Python objects to the majority of the built-in Python types. Serialization
17is the process of converting an object to a byte stream, and the inverse
18of which is converting a byte stream back to a Python object hierarchy.
20``dill`` provides the user the same interface as the ``pickle`` module, and
21also includes some additional features. In addition to pickling Python
22objects, ``dill`` provides the ability to save the state of an interpreter
23session in a single command. Hence, it would be feasible to save an
24interpreter session, close the interpreter, ship the pickled file to
25another computer, open a new interpreter, unpickle the session and
26thus continue from the 'saved' state of the original interpreter
27session.
29``dill`` can be used to store Python objects to a file, but the primary
30usage is to send Python objects across the network as a byte stream.
31``dill`` is quite flexible, and allows arbitrary user defined classes
32and functions to be serialized. Thus ``dill`` is not intended to be
33secure against erroneously or maliciously constructed data. It is
34left to the user to decide whether the data they unpickle is from
35a trustworthy source.
37``dill`` is part of ``pathos``, a Python framework for heterogeneous computing.
38``dill`` is in active development, so any user feedback, bug reports, comments,
39or suggestions are highly appreciated. A list of issues is located at
40https://github.com/uqfoundation/dill/issues, with a legacy list maintained at
41https://uqfoundation.github.io/project/pathos/query.
44Major Features
45==============
47``dill`` can pickle the following standard types:
49 - none, type, bool, int, float, complex, bytes, str,
50 - tuple, list, dict, file, buffer, builtin,
51 - Python classes, namedtuples, dataclasses, metaclasses,
52 - instances of classes,
53 - set, frozenset, array, functions, exceptions
55``dill`` can also pickle more 'exotic' standard types:
57 - functions with yields, nested functions, lambdas,
58 - cell, method, unboundmethod, module, code, methodwrapper,
59 - methoddescriptor, getsetdescriptor, memberdescriptor, wrapperdescriptor,
60 - dictproxy, slice, notimplemented, ellipsis, quit
62``dill`` cannot yet pickle these standard types:
64 - frame, generator, traceback
66``dill`` also provides the capability to:
68 - save and load Python interpreter sessions
69 - save and extract the source code from functions and classes
70 - interactively diagnose pickling errors
73Current Release
74===============
76The latest released version of ``dill`` is available from:
78 https://pypi.org/project/dill
80``dill`` is distributed under a 3-clause BSD license.
83Development Version
84===================
86You can get the latest development version with all the shiny new features at:
88 https://github.com/uqfoundation
90If you have a new contribution, please submit a pull request.
93Installation
94============
96``dill`` can be installed with ``pip``::
98 $ pip install dill
100To optionally include the ``objgraph`` diagnostic tool in the install::
102 $ pip install dill[graph]
104To optionally include the ``gprof2dot`` diagnostic tool in the install::
106 $ pip install dill[profile]
108For windows users, to optionally install session history tools::
110 $ pip install dill[readline]
113Requirements
114============
116``dill`` requires:
118 - ``python`` (or ``pypy``), **>=3.9**
119 - ``setuptools``, **>=42**
121Optional requirements:
123 - ``objgraph``, **>=1.7.2**
124 - ``gprof2dot``, **>=2022.7.29**
125 - ``pyreadline``, **>=1.7.1** (on windows)
128Basic Usage
129===========
131``dill`` is a drop-in replacement for ``pickle``. Existing code can be
132updated to allow complete pickling using::
134 >>> import dill as pickle
136or::
138 >>> from dill import dumps, loads
140``dumps`` converts the object to a unique byte string, and ``loads`` performs
141the inverse operation::
143 >>> squared = lambda x: x**2
144 >>> loads(dumps(squared))(3)
145 9
147There are a number of options to control serialization which are provided
148as keyword arguments to several ``dill`` functions:
150* with *protocol*, the pickle protocol level can be set. This uses the
151 same value as the ``pickle`` module, *DEFAULT_PROTOCOL*.
152* with *byref=True*, ``dill`` to behave a lot more like pickle with
153 certain objects (like modules) pickled by reference as opposed to
154 attempting to pickle the object itself.
155* with *recurse=True*, objects referred to in the global dictionary are
156 recursively traced and pickled, instead of the default behavior of
157 attempting to store the entire global dictionary.
158* with *fmode*, the contents of the file can be pickled along with the file
159 handle, which is useful if the object is being sent over the wire to a
160 remote system which does not have the original file on disk. Options are
161 *HANDLE_FMODE* for just the handle, *CONTENTS_FMODE* for the file content
162 and *FILE_FMODE* for content and handle.
163* with *ignore=False*, objects reconstructed with types defined in the
164 top-level script environment use the existing type in the environment
165 rather than a possibly different reconstructed type.
167The default serialization can also be set globally in *dill.settings*.
168Thus, we can modify how ``dill`` handles references to the global dictionary
169locally or globally::
171 >>> import dill.settings
172 >>> dumps(absolute) == dumps(absolute, recurse=True)
173 False
174 >>> dill.settings['recurse'] = True
175 >>> dumps(absolute) == dumps(absolute, recurse=True)
176 True
178``dill`` also includes source code inspection, as an alternate to pickling::
180 >>> import dill.source
181 >>> print(dill.source.getsource(squared))
182 squared = lambda x:x**2
184To aid in debugging pickling issues, use *dill.detect* which provides
185tools like pickle tracing::
187 >>> import dill.detect
188 >>> with dill.detect.trace():
189 >>> dumps(squared)
190 ┬ F1: <function <lambda> at 0x7fe074f8c280>
191 ├┬ F2: <function _create_function at 0x7fe074c49c10>
192 │└ # F2 [34 B]
193 ├┬ Co: <code object <lambda> at 0x7fe07501eb30, file "<stdin>", line 1>
194 │├┬ F2: <function _create_code at 0x7fe074c49ca0>
195 ││└ # F2 [19 B]
196 │└ # Co [87 B]
197 ├┬ D1: <dict object at 0x7fe0750d4680>
198 │└ # D1 [22 B]
199 ├┬ D2: <dict object at 0x7fe074c5a1c0>
200 │└ # D2 [2 B]
201 ├┬ D2: <dict object at 0x7fe074f903c0>
202 │├┬ D2: <dict object at 0x7fe074f8ebc0>
203 ││└ # D2 [2 B]
204 │└ # D2 [23 B]
205 └ # F1 [180 B]
207With trace, we see how ``dill`` stored the lambda (``F1``) by first storing
208``_create_function``, the underlying code object (``Co``) and ``_create_code``
209(which is used to handle code objects), then we handle the reference to
210the global dict (``D2``) plus other dictionaries (``D1`` and ``D2``) that
211save the lambda object's state. A ``#`` marks when the object is actually stored.
214More Information
215================
217Probably the best way to get started is to look at the documentation at
218http://dill.rtfd.io. Also see ``dill.tests`` for a set of scripts that
219demonstrate how ``dill`` can serialize different Python objects. You can
220run the test suite with ``python -m dill.tests``. The contents of any
221pickle file can be examined with ``undill``. As ``dill`` conforms to
222the ``pickle`` interface, the examples and documentation found at
223http://docs.python.org/library/pickle.html also apply to ``dill``
224if one will ``import dill as pickle``. The source code is also generally
225well documented, so further questions may be resolved by inspecting the
226code itself. Please feel free to submit a ticket on github, or ask a
227question on stackoverflow (**@Mike McKerns**).
228If you would like to share how you use ``dill`` in your work, please send
229an email (to **mmckerns at uqfoundation dot org**).
232Citation
233========
235If you use ``dill`` to do research that leads to publication, we ask that you
236acknowledge use of ``dill`` by citing the following in your publication::
238 M.M. McKerns, L. Strand, T. Sullivan, A. Fang, M.A.G. Aivazis,
239 "Building a framework for predictive science", Proceedings of
240 the 10th Python in Science Conference, 2011;
241 http://arxiv.org/pdf/1202.1056
243 Michael McKerns and Michael Aivazis,
244 "pathos: a framework for heterogeneous computing", 2010- ;
245 https://uqfoundation.github.io/project/pathos
247Please see https://uqfoundation.github.io/project/pathos or
248http://arxiv.org/pdf/1202.1056 for further information.
250'''
252__version__ = '0.4.1.dev0'
253__author__ = 'Mike McKerns'
255__license__ = '''
256Copyright (c) 2004-2016 California Institute of Technology.
257Copyright (c) 2016-2025 The Uncertainty Quantification Foundation.
258All rights reserved.
260This software is available subject to the conditions and terms laid
261out below. By downloading and using this software you are agreeing
262to the following conditions.
264Redistribution and use in source and binary forms, with or without
265modification, are permitted provided that the following conditions
266are met:
268 - Redistributions of source code must retain the above copyright
269 notice, this list of conditions and the following disclaimer.
271 - Redistributions in binary form must reproduce the above copyright
272 notice, this list of conditions and the following disclaimer in the
273 documentation and/or other materials provided with the distribution.
275 - Neither the names of the copyright holders nor the names of any of
276 the contributors may be used to endorse or promote products derived
277 from this software without specific prior written permission.
279THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
280"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
281TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
282PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR
283CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
284EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
285PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS;
286OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY,
287WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR
288OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF
289ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
291'''